Current approaches to the diagnosis of tuberculosis (TB) are not optimal. The tuberculin skin test has low sensitivity and specificity. Newer interferon-gamma release assays are more specific but lack sensitivity in some settings, are expensive, and require a relatively sophisticated laboratory. The microbiologic diagnosis of active TB disease is difficult in many patients for several reasons. In some, e.g. children and persons with HIV-TB co-infection, the low organism burden often limits the utility of sputum smear microscopy for diagnosis. Culture and nucleic acid-based diagnostic techniques are also limited by this low organism burden, as well as by the difficulties of sample handling and transport to central laboratories, especially in resource-limited settings. Illnesses with overlapping manifestations make clinical diagnosis of TB unreliable. Serologic tests are used to diagnose many infectious diseases, and have the advantages of high sensitivity and specificity, established methodologies, and adaptability to point of care test formats. Antibody-based tests for TB, however, using one or a few antigens, have been limited by inadequate sensitivity. New serologic tests that incorporate antigens that are specifically expressed during infection may have enhanced sensitivity and specificity. In vivo-expressed antigens also have the potential to identify disease stage-specific antibody responses. In addition to their potential for the development of new diagnostics, identification of disease stage-specific protein expression may provide insight into TB infection and disease pathogenesis. The overall objective of this research is to undertake a proteome-wide search for protein antigens of Mycobacterium tuberculosis (Mtb) that are recognized by the human humoral immune response. In the first aim of this project, we will synthesize nucleic acid programmable protein arrays (NAPPA) for approximately 3,000 Mtb proteins. In the second aim, these arrays will be probed with serum obtained at carefully defined stages of TB infection from Mtb-infected rabbits, an excellent model of human TB, including latent TB and reactivation after immune suppression. Screening of banked human serum from patients with well-documented TB will also be undertaken in this aim. These studies will identify candidate antigens to be analyzed in subsequent studies to screen for antigens that are broadly recognized by humans with TB, and antigens that provoke an antibody response at specific stages of TB infection. These data may lead to improved diagnostic tests for active and latent TB, and provide new insights into the pathogenesis of latent and active TB. PUBLIC HEALTH RELEVANCE: TB remains a major cause of illness and death, particularly in people living in resource-limited settings and persons who are HIV-infected. The proposed research is designed to provide the basis for new methods to diagnose TB. Earlier, simpler, and more sensitive and specific TB diagnosis could have a major impact on the identification and treatment of people with TB in the U.S. and worldwide.